The post-mortem pathology of Alzheimer's Disease is characterized by the presence in particular regions of the brain of many extracellular plaques and of many intracellular neurofibrillary tangles, whose density correlates with the severity of dementia. There is also massive, but regional, neuronal cell disfunction and cell loss, caused presumably by the reported neurotoxicity of the .beta.-amyloid peptides which are components of senile plaques. The cytotoxicity of the .beta.-amyloid peptides was first established in primary cell cultures from rodent brains and also in human cell cultures. These were relatively long-term experiments, lasting for a few days. The immediate molecular cause of the cytotoxicity was not clear from these reports. The work of Mattson et al. (J Neurosci. 12:376-389, 1992) indicates that .beta.-amyloid peptides, including the sequence .beta.AP.sub.25-35, in the presence of the excitatory neurotransmitter glutamate causes an immediate increase in intracellular calcium, which, it is supposed, is very toxic to the cell through its greatly increased second messenger activities.
The formation of pathological .beta.-amyloid peptides in Alzheimer's Disease is not well understood. The amyloid precursor protein (APP) is a very large transmembrane protein whose normal turnover degradation cleaves the presumptive .beta.-amyloid peptide in the middle, thus making it inactive as a neurotoxic agent. In addition, the future C-terminus of .beta.-amyloid peptides is buried in the middle of the lipid membrane. How the degradation of APP is altered in Alzheimer's Disease (AD) is only gradually becoming clear with no convincing explanation at present.
There are three .beta.-amyloid peptides, .beta.AP.sub.1-42, .beta.AP.sub.1-40, and .beta.AP.sub.25-35, which are homologous to the tachykinin neuropeptides. All three peptides are strongly neurotoxic when applied to cultured cells. .beta.AP.sub.1-40 and .beta.AP.sub.1-42 are the most prominent components of senile plaques. It is not clear whether .beta.AP.sub.25-35 occurs in the brains of AD individuals. .beta.AP.sub.25-35 might be absent because it has been scavenged when dead neurons are removed.
The .beta.AP.sub.1-42 peptide, and related shorter peptides, are cytotoxic towards cultured neuronal cells at micromolar concentrations, but neurotrophic at nanomolar concentrations. Others have observed that the peptide is cytotoxic also in vivo. Variability in results from different laboratories perhaps can be ascribed to the different propensities of particular .beta.-amyloid peptides to aggregate in aqueous solution. It has been suggested that long-term cytotoxicity resides in insoluble aggregates. The molecular mechanism of this cytotoxicity is not well known, perhaps because most of the reported experiments examine chronic cytotoxic effects only after 24-48 hours of exposure to insoluble aggregates of .beta.-amyloid peptides.
The ability of .beta.-amyloid peptides such as .beta.AP.sub.1-40 to form cation-selective ionophores was postulated earlier as a mechanism for cytotoxicity (Arispe et al., Proc. Nat'l Acad. Sci. USA 90:10573-10577, 1993; Arispe et al., Proc. Nat'l Acad Sci. USA 90:567-571, 1993). However, these experiments were carried out in artificial membranes. While in actual cells the ionophore mechanism might indeed be an important factor, there are at least two other mechanisms: interaction between the .beta.-amyloid peptides with existing ion channels, and penetration of the peptides into the cell with consequent release of calcium from internal stores.
Thus, while the precise mechanism of neurotoxicity of .beta.-amyloid peptides in Alzheimer's Disease has not been definitively established, there is a need to determine which of the aforementioned mechanisms of cytotoxicity is the cause of neuronal cell death in AD. Identification of the cytotoxic mechanism is needed to enhance the prospects of designing compounds capable of antagonizing the effects of aggregation of .beta.-amyloid peptides.